Reply to McIntyre and McKitrick: Proxy-based temperature reconstructions are robust

McIntyre and McKitrick (1) raise no valid issues regarding our paper. We specifically discussed divergence of “composite plus scale” (CPS) and “error-in-variables” (EIV) reconstructions before A.D. 1000 [ref. 2 and supporting information (SI) therein] and demonstrated (in the SI) that the EIV reconstruction is the more reliable where they diverge. The method of uncertainty estimation (use of calibration/validation residuals) is conventional (3, 4) and was described explicitly in ref. 2 (also in ref. 5), and Matlab code is available at www.meteo.psu.edu/∼mann/supplements/MultiproxyMeans07/code/codeveri/calc_error.m.